Top Banner
September 2008/1 Statis tics Canada Statisti que Canada From a seed to a forest The Research Data Centre Program
27

Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

Mar 28, 2015

Download

Documents

Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/1

Statistics Canada

Statistique Canada

From a seed to a forest

The Research Data Centre Program

Page 2: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/2

The Research Continuum(secondary sources)

Increasing level of detail

Page 3: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/3

What is the RDC Network?

It is a partnership that includes: More than 40 Canadian universities Major Granting Councils (SSHRC, CIHR,

CFI) Provincial governments Statistics Canada

Page 4: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/4

What is a Research Data Centre ?

Secure environment in a setting that is removed from Statistics Canada premises

Houses Statistics Canada micro data files Staffed by a Statistics Canada employee at all times Operates under the provisions of the Statistics Act Access limited to researchers with approved projects and

“sworn-in” under Statistics Act as “deemed employees” All researchers have direct access to the data

Page 5: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/5

General principle

Researchers from participating institutions do not pay for access to an RDC

Page 6: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/6

Toronto

COOL

U. Sask.

Moncton

SFU and

U. Victoria

Sudbury

Laval

McGill

Sherbrooke

UQAM

CIQSS

Toronto

Queen’s

U. Manitoba

UWO

COOL

Atlantic

CRISP

McMaster

SWORDC

Prairie

Alberta

BCIRDC

CIQSS

U. Windsor

York U.

Toronto

Page 7: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/7

Statistics Canada

Statistique Canada

Governance Structure

Page 8: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/8

RDC NETWORKUniversities, Centres &

Branches15 Data Centres 9 Branches26 Partners

Granting CouncilsSSHRC, CIHR

CFI

RDC National Coordinating Committee

(RDCNCC)

RDCNCC Chair

RDCNCC Office Assistant

Knowledge Transfer Officer

Statistics Canada

Regional Supervisors and RDC Analysts in

each Centre

Manager of RDC Network

STC RDC HQ Operations

Analysis & Dissemination

Unit

Page 9: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/9

Statistics Canada

Statistique Canada

How do researchers gain access to the RDCs

Page 10: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/10

Access to the Research Data Centres(Academic researchers)

Project proposal

Proposal evaluation - SSHRC

Security clearance - enhanced reliability check

Orientation session and “oath of office”

Researcher agrees to provide publicly available report that falls within Statistics Canada’s mandate

Page 11: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/11

Evaluation Criteria – academic researchers Necessity for access to detailed micro data and to the

particular dataset(s) specified in proposal Clearly defined project objectives Suitability of analytical and statistical methods Scientific merit of the project Do the applicant and the team members (if applicable)

have the experience, qualifications, and expertise to successfully complete the proposed project?

Page 12: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/12

Access to the Research Data Centres(Federal/Provincial/Territorial researchers)

Project proposal, “reviewed” by designated official (either Director General research for federal departments, or statistical focal point for provinces and territories)

The work is to be performed by an employee (or contract employee) of the federal department or the provincial/territorial government

Proposal evaluation - STC

Security clearance - enhanced reliability check

Orientation session and “oath of office”

Page 13: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/13

Evaluation Criteria – federal, provincial, territorial researchers The work is statistical in nature. The work requires access to the specified data file(s),

and cannot be accomplished using publicly available data files or published statistical information.

The work can be completed using the specified data file(s) within the time frame specified.

The work can not be quickly and efficiently completed by Statistics Canada employees.

No data sharing agreement exists within the requesting department with respect to the dataset for which access is required.

Page 14: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/14

Statistics Canada

Statistique Canada

Current status

Page 15: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/15

The statistics

0

200

400

600

800

1000

1200

1400

Projects

Cummulative to March 31, 2008

Withdrawn

Active

Completed

Withdrawn 1 8 20 38 51 70 70

Active 14 71 181 280 373 449 572 703 743

Completed 0 0 12 46 125 249 374 517 568

2000 2001 2002 2003 2004 2005 2006 2007 20080

200

400

600

800

1000

1200

1400

1600

1800

2000

No. of researchers as of March 31, 2008

Student

Non-student

Student 17 26 179 239 280 394 461 648 616

Non-student 53 259 248 327 320 626 1031 1310 1277

2000 2001 2002 2003 2004 2005 2006 2007 2008

Page 16: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/16

The statistics (cumulative to March 31, 2008)

25 operating centres, and growing! Secured a five year operating grant from

SSHRC and CIHR, starting in 2005 Secured two infrastructure grants from CFI

totalling over $10 million Six annual thematic conferences based on

research conducted in the RDCs

Page 17: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/17

http://www.statcan.ca/english/rdc/

Page 18: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/18

Statistics Canada

Statistique Canada

Data holdings

Page 19: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/19

A sample of the data that are in the RDCs

Aboriginal Peoples Survey (APS) 

Canadian Community Health Survey (CCHS) Cycle 3.1 Cycle 2.2 - Nutrition Cycle 2.1 Cycle 1.2 - Mental Health and Well-being Cycle 1.1

 Census of Population2001 Census 1996 Census1991 Census

Ethnic Diversity Survey (EDS)

General Social Survey (GSS) Access to and Use of Information Communication Technology Education, Work and Retirement Family Health Social Engagement Social Support and Aging Time Use Victimization

Longitudinal Survey of Immigrants to Canada (LSIC)

National Graduates Survey (NGS) 

National Longitudinal Survey of Children and Youth (NLSCY)

National Population Health Survey (NPHS) Household Component - Cross-sectional Household Component - Longitudinal North Component Health Institutions Component

Participation and Activity Limitation Survey (PALS)

Survey of Labour and Income Dynamics (SLID)

Workplace and Employee Survey (WES)

Youth in Transition Survey (YITS)

Program for International Student Assessment (PISA)   

Page 20: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/20

Differences between public files and detailed master files in the RDCsPublic files RDC master files

Level of geography = province or CMA

Census Subdivision,

Census Tract or below

Aggregates certain countries of birth or ethnic origins

All the ethno-cultural details are available

Joint analysis at individual and family level is limited

Master files can be used with full individual level information and characteristics of families

Only cross-sectional data Panel data tracking the same respondents over time (not for Census data)

Census public file is only a sample The Census master file with over 6 millions records is available

Page 21: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/21

Statistics Canada

Statistique Canada

Ensuring security and confidentiality

Page 22: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/22

What are the security measures and how are the data protected?

The legal frameworklegal framework for access: research contract, security clearance and oath of office

Physical protectionPhysical protection of data: controlled access to the labs, individual computer accounts, limited network access

Analytical protectionAnalytical protection of data: a process of disclosure avoidance control is applied for each request to release statistical output

Analytical protectionAnalytical protection of data (part 2): minimum number of observations in each cell for descriptive (tabular) output

Page 23: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/23

Specific disclosure avoidance rules for Census data

Special rules are necessary considering the coverage of the census, hence the higher risk of identifying respondents with rare characteristics

Random roundingRandom rounding of tabular and descriptive output

Minimum populationMinimum population thresholds for the smallest unit of geography in the analysis

Release of intermediate output is discouraged Risk of residual disclosure Increased workload associated with verifying the

results against disclosure

Page 24: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/24

Statistics Canada

Statistique Canada

Challenges and future directions

Page 25: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/25

Today’s Challenges

Add to the Census data holdings in the RDCs Add administrative data to the holdings in the

RDCs Coordination with different modes of access Access to international data and by international

researchers Continued growth of the network Publication of research Link all the RDCs by a secure wide area network

Page 26: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/26

Major initiatives for the future

Add administrative data linked to survey data

Add historical census data Fully document the data holdings using

DDI compliant technology

Page 27: Statistics Canada Statistique Canada September 2008/1 From a seed to a forest The Research Data Centre Program.

September 2008/27

To learn more …

http://rdc-cdr.net/

http://www.statcan.ca/francais/rdc/

http://www.statcan.ca/english/rdc/

[email protected]

The endThe end